Re: Filters & Chokes

["Ed Phillips" <evp-at-xxxxxxxxxxx> (by way of Terry Fritz <twftesla-at-xxxxxxxxxx>)]

Tesla list wrote:
> 
> Original poster: "Lau, Gary" <Gary.Lau-at-compaq-dot-com>
> 
> I use a standard unmodified 15/60 NST and the filter uses a pair of 1.6K
> 113W resistors.  For a very long time I was puzzled at just how hot these
> got.  By standard logic, the power is I^2 R, or .06^2 * 1600 = 5.76Watts.
> But after building a fiber-optic VI probe, I was able to scope the NST
> secondary current with a digital scope, which also performs true-RMS
> calculations.  Guess what?  My 60mA NST is pushing 184 mA, RMS.  It appears
> that some unexpected things happen when charging an LTR cap, speculation is
> that the current shunts saturate and allow far more current to pass.  So the
> 184 mA causes a dissipation of 54 Watts per resistor.
> 
> Gary Lau
> Waltham, MA USA	

	No mystery here.  When you hook the capacitor to the secondary of an
NST the available current goes up, since the negative reactance
"cancels" some of the positive reactance built into the transformer for
current limiting and you no longer have a "60 ma transformer". (By the
way, it's not a case of shunt saturation.)  In fact, at the magic
"matched capacitance" value the leakage reactance is effectively
cancelled (ignoring tolerances, changes in permeability with flux
density, etc.) and you should be able to draw current limited only by
the resistance of the transformer.  In practice, the thing would fail
from insulation breakdown, as the 60 cycle Q of a typical transformer
(measured here at low voltage) is of the order of 12.  I suspect this
resonant voltage rise is the source of most NST failures when guys get
greedy and open up the gap too much.

Ed